The major aim of this study is to assess the risk and modification of risk of breast cancer and benign breast disease with atypical breast fluid cytologic findings in a cohort of 5206 women who participated in previous breast fluid studies between 1970 and 1980. In this cohort (the only one of its kind), 425 women had atypical breast fluid cytologic findings (moderate or atypical hyperplasia), 2769 women had cytologic findings that were not atypical (normal and mild hyperplasia), 607 women yielded breast fluid specimens that were unsatisfactory for cytologic diagnosis, and breast fluid collection was unsuccessfully attempted from 1405 women (nonsecretors). Epidemiologic data was collected for all women. To determine whether or not each women in the cohort developed benign or malignant breast disease we will contact each women or a next of kin by 1) mail (or telephone) using most current address, updated current or death address from "2-4" below, California Department of Motor Vehicles, or inquiries to former neighbors, referring physicians or other sources; 2) California Tumor Registry 3) California Automated Mortality Linkage System and National Death Index; and 4) National Cancer Institute Breast Screening Center follow-up data. All reports of benign and malignant breast disease will be validated at a minimum by review of pathology reports and all pathology and specimens will be sought for standardized review. This study has 80% power to detect a 2.5-fold risk of breast cancer associated with cytologic findings of atypia compared to no atypia, with only 60% follow-up (alpha = .05) . By quantifying the risks and modification of risk associated with breast fluid secretion and cytologic classification, the study will provide new and valuable information for detecting women at high risk breast cancer and may detect factors that modify risk.